https://llama.meta.com/llama3/

- There's a new tool called file_search that can handle 10,000 context and data files per assistant, which is considerably more than before. Plus, it's faster and smarter with searches. - Vector store objects are a new add. You drop a file in, and it gets set up for quick searches. Helps keep things organized and cuts down on embedding and vector DB hassle. - You've got more control over how many tokens the system uses each time, so you can keep an eye on costs. - There's a tool_choice option now. It lets you pick exactly which tool you want to use for each task. - You can make messages that look like they're from the assistant itself—handy for setting up conversations the way you want. - The setup for assistants and runs has more options, like tweaking how creative or precise the responses are. - Only the newer, custom-tuned versions of the GPT models are available to use. - The whole system supports streaming now, so it's smoother and more like real-time. - They’ve added some tools in their software kits to help manage streaming and checking for updates. All in all, I think this is a great step in the right direction.

I am Spencer Bentley, an AI Systems Engineer from the UK and a senior member of the leadership and management team for the OpenAI developer forum. It’s great to be a part of this group, which I hope will bring together top AI developers, automation agencies, and innovative companies to drive AI adoption forward. I’ve seen firsthand the incredible growth and impact of AI technologies. The OpenAI forum has over 600,000 members, nearly a million posts, over 12k active weekly users and sees an average of around a thousand a day in new sign-ups. It's a testament to the growing interest and importance of AI in today's world. Over the past several years, I've enjoyed discussions and collaborations on cutting-edge projects. I’ve read over 100 thousand posts to get a feel for the undercurrent in the AI developer community. AI Systems Engineering is one of the most important fields in this developing technology, I am particularly interested in the intersection of AI and automation. I believe that AI-powered automation agencies have an influential and profitable roll to play to bring industries into the new epoch and drive significant efficiency gains. I look forward to discussing the latest advancements, real-world applications, and best practices you all. In this group, I hope to foster an in-depth technical discussion environment, share practical insights, and collaborate with like-minded devs and entrepreneurs to push what's possible with AI and automation. Whether you're a seasoned developer, a forward-thinking agency, or an innovative company. I'm looking forward to engaging with you all, learning from your experiences, and working together to shape how we build AI automation. If you have any questions, ideas, or just want to connect, please don't hesitate to reach out.

Write a quick introduction about yourself: - Where you're from - What you're working on

https://platform.openai.com/docs/assistants/whats-new Excellent set of new features including a 10k file limit on assistants file retrievals. While I still prefer to use traditional code to build RAG pipelines, the OpenAI Assistants API can be a powerful tool for low-code builders and AAA practitioners to rapidly deploy solutions at scale. I'll be taking a closer look at the new features in an upcoming post. One of the most requested features so far has been the ability to control the context size in Assistants' retrievals, it will be interesting to see if this has been addressed to developers liking in the new API.